Production risk identification system of automatic battery assembly production line

Abstract

The invention discloses a production risk identification system of an automatic battery assembly production line, and relates to the field of risk identification, and the system comprises a management module which serves as a central control end of each functional module and unit, is in butt joint with a bus system of the assembly production line, and obtains data access authority and instruction issuing authority; the assembly acquisition module is used for acquiring the operation duration, the fault frequency, the downtime and the operation efficiency of each assembly line device through a bus system and a preset sensor to form an operation data set; the quality detection module is used for acquiring and analyzing assembly product quality data in a production cycle in real time and generating a quality score for each assembly product; the operation data is analyzed and predicted, dynamic simulation is carried out in the production process, the production influence coefficient of a future period is predicted, risks are recognized in advance, the production efficiency is ensured, meanwhile, the optimal opportunity for equipment intervention is provided, and intelligent maintenance management is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of risk identification, and in particular to a production risk identification system for a battery automatic assembly production line. Background Art

[0002] With the rapid development of electric vehicles and renewable energy, the demand for batteries has surged. In particular, the production of lithium-ion batteries has become an industry focus. Modern manufacturing is moving towards high automation and intelligence. Companies are building intelligent production lines to improve production efficiency and reduce human errors. Battery production involves multiple components and steps. The assembly process is complex and sensitive to the environment and equipment status. Any small deviation may lead to product quality problems. Therefore, a system is needed to monitor and identify potential risks in real time.

[0003] With the maturity of big data and artificial intelligence technologies, the manufacturing industry is increasingly relying on data-driven decision-making. Production risk identification systems need to integrate and analyze large amounts of historical and real-time data to provide accurate predictions and recommendations to help companies respond quickly. Furthermore, as market competition becomes increasingly fierce, companies need to enhance their competitiveness by improving production efficiency and product quality.

[0004] However, existing production risk identification systems fail to fully utilize interactions and multivariate analysis, resulting in insufficient accuracy in identifying production risks, prone to missed or misjudgment, and difficult to dynamically adjust according to real-time production conditions, resulting in inflexible production planning or scheduling. They only focus on a single variable, such as monitoring equipment failures, while ignoring the interactions between multiple variables. This leads to insufficient risk assessment of the overall production process and an inability to effectively determine when to perform preventive intervention on equipment, resulting in premature or late maintenance of equipment, affecting production efficiency. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides a production risk identification system for a battery automatic assembly production line, which can effectively solve the problems of the prior art.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions.

[0007] The present invention discloses a production risk identification system for a battery automatic assembly production line, comprising:

[0008] Management module, used to connect to the bus system of the assembly production line and obtain data and instruction permissions;

[0009] The assembly collection module is used to collect the operating time, number of failures, downtime and operating efficiency of each assembly line equipment to form an operating data set;

[0010] The quality inspection module is used to acquire and analyze the quality data of assembly products during the production cycle in real time and generate quality scores;

[0011] Supply access module, used to obtain and update the completion progress of production orders in real time;

[0012] The assembly simulation module is used to construct an assembly model that simulates the production impact coefficient of the current product in the assembly process by using historical equipment operation data, assembly progress, assembly product quality score, and production order completion status as samples;

[0013] The impact prediction module is used to obtain the trained assembly model through the assembly simulation module, input the current operation data of a certain equipment, assembly progress, assembly product quality score and production order completion status, and predict and output the production impact coefficient of the equipment on the production order completion status and assembly product quality score in the next few cycles;

[0014] A determination module is used to preset a standard production impact coefficient for the current cycle, and based on the standard production impact coefficient, evaluate the production impact coefficients of several cycles output by the impact prediction module, and mark the cycles whose production impact coefficients exceed the preset standard production impact coefficient;

[0015] The joint analysis unit is used to comprehensively analyze several cycles before the cycle marked by the judgment module, combine the current order status coefficient and the equipment life impact coefficient, and output a recommended cycle for starting intervention;

[0016] The stage identification module is used to perform real-time detection within the intervention period recommended by the joint analysis unit. According to the preset period, it obtains the actual expression of the production impact coefficient in stages, judges one by one whether there is any discrepancy between the production impact coefficient of the corresponding period and the preset limit, and creates a smart contract on the blockchain to automatically trigger the contract terms adjustment intervention measures.

[0017] Furthermore, the management module is interactively connected to a case storage module via a wireless network. The case storage module is used to store historical cases of operating data of production influencing parameters, assembly progress, assembly product quality scores and production order completion status parameters through customized verification, and mark and classify the stored historical cases as a sample library for the assembly simulation module.

[0018] Furthermore, the calculation formula of the production impact coefficient output by the impact prediction module is:

[0019]

[0020] In the formula, P(t+i) represents the production impact coefficient in the time period t+i, T0 represents the intercept term of the model, and a j Represents the weight of the input variable, Xj (t) represents the jth device operation data in time period t, n represents the number of input variables, b m Represents the weight of the assembly progress variable, U m (t) represents the progress coefficient of the current assembly work, P represents the number of assembly progress variables, c l represents the weight of the product quality score, d represents the weight of the production order status, Q l (t) represents the current production order completion status, A(t) represents the impact of equipment aging or wear on production capacity, C(t,i) represents the impact of equipment at time t on the future i cycles, V i represents the random influence coefficient related to the time period, e represents the weight of the equipment life influence coefficient, and f represents the weight of the interaction influence.

[0021] Furthermore, the working logic of the calculation formula of the production impact coefficient is:

[0022] Collect all input parameters related to the prediction of production impact coefficients for several future periods;

[0023] Use regression analysis to calculate model parameters, and use the least squares method to minimize the error between the predicted value and the actual value to obtain the optimal parameters;

[0024] The model parameters are determined, and the required input data are collected in each cycle to calculate the production impact coefficients for several future cycles.

[0025] Furthermore, the joint analysis unit is deployed with submodules at the lower level, including: an order extraction module, a machine analysis module, a cross-identification module, and an opinion output module. The order extraction module and the machine analysis module are interactively connected to the cross-identification module via a wireless network, and the cross-identification module is interactively connected to the opinion output module via a wireless network, wherein:

[0026] The order extraction module is used to extract order information of several periods before the period marked by the determination module through the impact prediction module, including order quantity, delivery date and current completion progress, and output the order status coefficient;

[0027] The machine analysis module is used to obtain real-time equipment operation data and fault data through the assembly acquisition module, and use pre-set performance indicators to evaluate the equipment's operating status for several cycles, and output the equipment life impact coefficient for several cycles before the cycle marked by the judgment module;

[0028] The cross-identification module is used to integrate the order status coefficients of several cycles provided by the order extraction module with the equipment life impact coefficients of several cycles provided by the machine analysis module, compare the production impact coefficient of each cycle with the preset standard production impact coefficient, and identify the fluctuation status of the two types of coefficients;

[0029] The opinion output module is used to summarize the recognition results of the cross-recognition module in several periodic fluctuation states, sort them from high to low priority, and output the period with the highest priority as the recommended intervention period;

[0030] The cross-identification module synchronously receives feedback data from the order extraction module and the machine analysis module, and performs attribute matching of the two types of feedback data within the current cycle to form several data structures with the same attributes as initial integration references.

[0031] Furthermore, the calculation formula for the cross-recognition module to identify two types of coefficient fluctuation states is:

[0032]

[0033] Where R i Represents the order status coefficient OS in the i-th period i and equipment life impact coefficient EL i Relative to their respective target values OS target and EL target The relative difference in OS i Represents the order status coefficient of the i-th period, OS target Represents the target standard value of the order status coefficient, EL i Represents the equipment life impact coefficient of the i-th cycle, EL target Represents the target standard value of the equipment life impact coefficient;

[0034] The calculation formula of the two types of coefficient fluctuation states is obtained by R i , perform statistical analysis and calculate the average value of the overall fluctuation state. If the average value exceeds a certain preset threshold, it is considered that there is an obvious fluctuation state.

[0035] Furthermore, the working logic of the stage identification module to create a smart contract on the blockchain for intervention and automatically trigger the contract terms adjustment intervention measures is as follows:

[0036] The contract terms set the monitoring period, production conditions and intervention conditions. During the phase change, when an actual production parameter exceeds the set intervention threshold, the smart contract will automatically trigger one or more intervention measures, including:

[0037] Each stage change will generate an unalterable hash value through real-time recording and storage in the blockchain. Based on the actual expression data of the collected production impact coefficient, contextual data including the equipment's operating status, environmental parameters and quality inspection results will be generated;

[0038] Smart contracts deployed on the blockchain will be adjusted accordingly based on real-time situational data, triggering different contract terms according to situational changes. When a preset specific situation is identified, the execution parameters of the smart contract will be dynamically changed based on custom rules and serve as a reference for adjustments to the joint analysis unit.

[0039] Furthermore, the determination module and the stage identification module are connected to an alarm module via an electrical medium. The alarm module is used to receive the determination results of the determination module and the stage identification module in real time. When any judgment result is negative, an alarm behavior feedback is immediately provided to the management module.

[0040] Furthermore, the stage identification module is interactively connected to a patrol module via a wireless network. The patrol module is used to monitor the changes in the production impact coefficient after the intervention of the joint analysis unit in real time, evaluate the identification results of the stage identification module, and synchronously feed back to the management module.

[0041] Furthermore, the management module is interactively connected to the assembly acquisition module, the quality inspection module and the supply access module through a wireless network, the assembly simulation module is interactively connected to the assembly acquisition module, the quality inspection module, the supply access module and the impact prediction module through a wireless network, the impact prediction module is interactively connected to the determination module through a wireless network, the determination module is interactively connected to the joint analysis unit through a wireless network, and the joint analysis unit is interactively connected to the stage identification module through a wireless network.

[0042] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0043] 1. By collecting and monitoring operation data and product quality in real time, and building assembly models to analyze and predict operation data, dynamic simulation is performed during the production process to predict the production impact coefficient of future cycles. By combining historical data and multivariate analysis, the potential impact of production progress and quality is evaluated in real time, and risks are identified in advance. Through comprehensive analysis of multiple factors, key influencing factors can be identified more accurately, providing a basis for optimizing production decisions.

[0044] 2. Through the joint analysis module, the status and influencing parameters of multiple cycles before the marking cycle are comprehensively analyzed. While ensuring production efficiency, the best time to intervene in the equipment can be proposed, intelligent maintenance management can be realized, the service life of the equipment can be extended, and production continuity can be improved. It has strong adaptability and flexibility, and can judge in real time whether there are situations that exceed the preset limits in the production process, respond to changes in time, and improve coping capabilities. In this way, while ensuring full production efficiency, it avoids a major impact on order completion and equipment life, thereby helping users to maximize the operating production capacity of existing mechanical equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0046] Figure 1 This is a schematic diagram of the framework of a production risk identification system for a battery automatic assembly production line in the present invention;

[0047] Figure 2 Schematic diagram of the framework of the joint analysis unit in the present invention.

[0048] The numbers in the figure represent the following: 1. Management module; 2. Assembly acquisition module; 3. Quality inspection module; 4. Supply access module; 5. Assembly simulation module; 6. Impact prediction module; 7. Judgment module; 8. Joint analysis unit; 81. Order extraction module; 82. Machine analysis module; 83. Cross-recognition module; 84. Opinion output module; 9. Stage identification module; 10. Alarm module; 11. Patrol module; 12. Case storage module. DETAILED DESCRIPTION

[0049] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0050] The present invention will be further described below with reference to the embodiments.

[0051] Example 1

[0052] This embodiment is a production risk identification system for a battery automatic assembly production line, such as Figure 1 As shown, including:

[0053] Management Module 1, serving as the central control terminal for all functional modules and units, connects to the bus system of the assembly line to obtain data access rights and command issuance authority. It integrates all functional modules and units to achieve centralized management, enabling rapid acquisition of production data and issuance of commands, improving decision-making efficiency.

[0054] The management module 1 is interactively connected to the case storage module 12 via a wireless network. The case storage module 12 is used to store historical cases of operating data, assembly progress, assembly product quality scores, and production order completion status parameters that are verified through customized production, and to mark and classify the stored historical cases. As a sample library for the assembly simulation module 5, the introduction of the case storage module 12 can effectively accumulate and utilize historical cases, optimize the model training process, improve the accuracy of model predictions, and provide historical data support for subsequent decision-making and system improvements, forming a virtuous cycle of intelligent management system.

[0055] Assembly collection module 2, used to collect the operating time, number of failures, downtime and operating efficiency of each assembly line equipment through the bus system and preset sensors to form an operating data set;

[0056] Quality inspection module 3, used to acquire and analyze assembly product quality data during the production cycle in real time and generate a quality score for each assembly product;

[0057] Supply Access Module 4 is used to connect to the order management system and warehouse management system to obtain and update the completion progress of production orders in real time. By introducing Supply Access Module 4, comprehensive monitoring of the completion status of production orders can be achieved, providing more solid data support for production decision-making, thereby effectively improving the flexibility and responsiveness of the production line, helping to optimize resource allocation and maximize production efficiency while meeting customer needs.

[0058] Assembly simulation module 5 is used to construct an assembly model that simulates the production impact coefficient of the current product in the assembly process by using historical operation data, assembly progress, assembly product quality score and production order completion status of a certain equipment as samples;

[0059] Impact prediction module 6 is used to obtain the trained assembly model through assembly simulation module 5, input the current operation data of a certain equipment, assembly progress, assembly product quality score and production order completion status, and predict and output the production impact coefficient of the equipment on the production order completion status and assembly product quality score in the next several cycles;

[0060] The determination module 7 is used to preset a standard production impact coefficient for the current cycle. Based on the standard production impact coefficient, it evaluates the production impact coefficients of several cycles output by the impact prediction module 6 and marks the cycles whose production impact coefficients exceed the preset standard production impact coefficients. The determination module 7 and the stage identification module 9 are connected to the alarm module 10 via an electrical medium. The alarm module 10 is used to receive the judgment results of the determination module 7 and the stage identification module 9 in real time. When any judgment result is negative, the alarm behavior feedback is immediately sent to the management module 1.

[0061] The joint analysis unit 8 is used to comprehensively analyze several cycles before the cycle marked by the determination module 7, combine the current order status coefficient and the equipment life impact coefficient, and output a recommended cycle for starting intervention;

[0062] The joint analysis unit 8 has submodules deployed at the lower level, including: an order extraction module 81, a machine analysis module 82, a cross-identification module 83, and an opinion output module 84. The order extraction module 81 and the machine analysis module 82 are interactively connected to the cross-identification module 83 via a wireless network, and the cross-identification module 83 is interactively connected to the opinion output module 84 via a wireless network, wherein:

[0063] The order extraction module 81 is used to extract order information of several periods before the period marked by the determination module 7 through the impact prediction module 6, including order quantity, delivery date and current completion progress, and output the order status coefficient;

[0064] The machine analysis module 82 is used to obtain real-time equipment operation data and fault data through the assembly acquisition module 2, and evaluate the equipment operation status for several cycles using pre-set performance indicators, and output the equipment life impact coefficient for several cycles before the cycle marked by the determination module 7;

[0065] The cross-identification module 83 is used to integrate the order status coefficients for several cycles provided by the order extraction module 81 with the equipment life impact coefficients for several cycles provided by the machine analysis module 82, compare the production impact coefficients for each cycle with the preset standard production impact coefficients, and identify the fluctuation status of the two types of coefficients. The cross-identification module 83 synchronously receives feedback data from the order extraction module 81 and the machine analysis module 82, and performs attribute matching on the two types of feedback data within the current cycle to form several data structures with the same attributes as a reference for initial integration.

[0066] The opinion output module 84 is used to summarize the identification results of the cross identification module 83 in the fluctuation state of several cycles, sort them from high to low priority, and output the cycle with the highest priority as the recommended intervention cycle;

[0067] The stage identification module 9 is used to perform real-time detection within the intervention period recommended by the joint analysis unit 8. According to the preset period, the actual expression of the production impact coefficient is obtained in stages. The production impact coefficient of the corresponding period is judged one by one to see whether there is any discrepancy exceeding the preset limit. A smart contract is created on the blockchain, and the contract terms of the monitoring period, production conditions and intervention conditions are preset. During the stage change stage, when an actual production parameter exceeds the set intervention threshold, the smart contract will automatically trigger one or more intervention measures, including:

[0068] Each stage change will generate an unalterable hash value through real-time recording and storage in the blockchain. Based on the actual expression data of the collected production impact coefficient, contextual data including the equipment's operating status, environmental parameters and quality inspection results will be generated;

[0069] Smart contracts deployed on the blockchain will be adjusted accordingly based on real-time contextual data, triggering different contract terms based on contextual changes. When a specific preset context is identified, the execution parameters of the smart contract will be dynamically changed based on custom rules, and serve as a reference for adjustments in the joint analysis unit 8. By combining contextual awareness technology with blockchain, an adaptive, highly transparent, and traceable stage identification and verification mechanism is formed, greatly improving the intelligence level of the battery automatic assembly production line. It can sense environmental changes in real time and automatically adjust contracts and their execution to ensure the safety and efficiency of the production process. In addition, through an explainable decision-making mechanism and contextual self-learning capabilities, the long-term iterative optimization of the system is ensured, providing enterprises with a strong competitive advantage.

[0070] The stage identification module 9 is interactively connected to the patrol module 11 via a wireless network. The patrol module 11 is used to monitor the changes in the production impact coefficient after the intervention of the joint analysis unit 8 in real time, evaluate the identification results of the stage identification module 9, and synchronously feed back to the management module 1;

[0071] like Figure 1 As shown, the management module 1 is interactively connected with the assembly acquisition module 2, the quality inspection module 3, and the supply access module 4 via a wireless network; the assembly simulation module 5 is interactively connected with the assembly acquisition module 2, the quality inspection module 3, the supply access module 4, and the impact prediction module 6 via a wireless network; the impact prediction module 6 is interactively connected with the determination module 7 via a wireless network; the determination module 7 is interactively connected with the joint analysis unit 8 via a wireless network; and the joint analysis unit 8 is interactively connected with the stage identification module 9 via a wireless network;

[0072] During implementation, this embodiment uses the management module 1 to control the global function modules, the assembly collection module 2 to identify the operating data of a certain assembly equipment, the quality inspection module 3 to obtain the quality score of the produced products, the supply access module 4 to obtain the product supply chain data, the assembly simulation module 5 to build the assembly model, and the impact prediction module 6 to obtain the trained assembly model. The operating time, number of failures, downtime, and operating efficiency of the current equipment operation data are input to simulate the current assembly progress, and the production impact coefficient on the completion status of the production order and the quality of the assembled products in each of the future cycles is predicted and output, thereby helping users to maximize the operating and production capacity of existing machinery.

[0073] The determination module 7 determines the production impact coefficients of the obtained cycles, marks the cycles with the impact coefficient exceeding the threshold, and conducts a comprehensive analysis of the cycles before the marked cycle by the joint analysis unit 8. The comprehensive analysis content includes: the current order completion status and the parameters affecting the life of the assembly equipment. The cycle with the lowest comprehensive impact is jointly output, and it is recommended to intervene in the assembly equipment in this cycle, so as to avoid significant impact on order completion and equipment life while ensuring sufficient production efficiency;

[0074] Through the stage identification module 9, the actual expression of the production impact coefficient of the actual production cycle is obtained in stages according to the preset period, and whether there is a discrepancy between the predicted data and the actual data is judged in stages. The judgment results of the judgment module 7 and the stage identification module 9 are fed back through the alarm module 10, and the recognition results of the stage identification module 9 are evaluated and fed back through the patrol module 11. The historical collected data and analysis data are stored through the case storage module 12 as training samples for the assembly simulation module 5.

[0075] Compared with existing technologies, this system collects and analyzes key data from the production process in real time, providing strong data support for production decisions. The introduction of historical case storage and model training enables continuous system optimization and improves prediction accuracy. Through a virtuous cycle of intelligent management, it helps to quickly adapt to changes and improve production efficiency.

[0076] Real-time monitoring of key indicators in the production process, and immediate feedback of alarms when indicators exceed preset standards, thereby achieving rapid response and intervention to ensure that customer needs can be met. By analyzing historical operating data, potential risks are identified and recommended intervention cycles are output to achieve early warning and intervention of problems, thereby reducing production risks, enabling production to be adjusted according to actual needs, and optimizing resource allocation.

[0077] Example 2

[0078] At other levels, such as Figure 2As shown, this embodiment also provides a calculation formula for the production impact coefficient, which is specifically:

[0079]

[0080] In the formula, P(t+i) represents the production impact coefficient in the time period t+i, T0 represents the intercept term of the model, which indicates the basic level of the impact coefficient, and a j Represents the weight of the input variable, which is related to the j-th input variable X j The associated regression coefficient indicates the degree of influence of the input variable on the production impact coefficient, X j (t) represents the jth device operation data within the time period t, reflecting the operating status of the device, n represents the number of input variables, b m Represents the weight of the assembly progress variable, which is the weight of the mth assembly progress variable U m The relevant regression coefficient represents the impact of assembly progress on production impact coefficient, U m (t) represents the progress coefficient of the current assembly work, P represents the number of assembly progress variables, c l Represents the weight of the product quality score, which is related to the lth product quality score variable Q l The relevant regression coefficient represents the impact of product quality on the production impact coefficient. d represents the weight of the production order status and is the regression coefficient associated with the current production order status O, indicating the degree of influence of the order status on production. Q l (t) represents the current production order completion status, A(t) represents the impact of equipment aging or wear on production capacity, C(t,i) represents the impact of the equipment at time t on the future i cycles, which is the effect of the equipment's working status at a certain point in time on subsequent production capacity, and V i represents the random influence coefficient related to the time period. In order to capture the fluctuation of periodic changes, e represents the weight of the equipment life influence coefficient. It is the regression coefficient related to the equipment life influence coefficient A(t), reflecting the influence of equipment aging or wear on the production influence coefficient. f represents the weight of the interaction influence. It is the regression coefficient related to the interaction influence C(t,i) between time t and future period i, indicating the dynamic change of the impact of the current status of the equipment on future production.

[0081] The working logic of the calculation formula of the production impact coefficient is:

[0082] Collect all input parameters related to the prediction of production impact coefficients for several future periods;

[0083] Use regression analysis to calculate model parameters, and use the least squares method to minimize the error between the predicted value and the actual value to obtain the optimal parameters;

[0084] The model parameters are determined, and the required input data are collected in each cycle to calculate the production impact coefficients for several future cycles.

[0085] This formula takes into account multiple key factors such as the order status coefficient and the equipment life impact coefficient. It can fully reflect the comprehensive effect of different variables on the production impact coefficient during the production process. Through the design of multi-variable calculation, the system can more accurately reflect the actual situation.

[0086] Example 3

[0087] In this embodiment, two calculation formulas for coefficient fluctuation states are provided, specifically:

[0088]

[0089] The calculation formula of the two types of coefficient fluctuation states is obtained i , perform statistical analysis and calculate the average value of the overall fluctuation state. If the average value exceeds a certain preset threshold, it is considered that there is an obvious fluctuation state;

[0090] Where R i Represents the order status coefficient OS in the i-th period i and equipment life impact coefficient EL i Relative to their respective target values OS target and EL target The relative difference in OS i Represents the order status coefficient of the i-th period, OS target Represents the target standard value of the order status coefficient, EL i Represents the equipment life impact coefficient of the i-th cycle, EL target Represents the target standard value of the equipment life impact coefficient;

[0091] This formula can quickly calculate the fluctuation status of each cycle and achieve real-time monitoring through batch calculations, thereby providing timely feedback to management. The real-time updated data helps decision makers quickly identify possible problems in the system and take corresponding measures.

[0092] In summary, this invention collects equipment operation data in real time, builds and trains an assembly simulation model, simulates the assembly process under different conditions, and predicts production impact coefficients for multiple future cycles. This helps users understand how different factors affect order fulfillment and product quality, identifies and marks potential risk cycles, provides recommendations for subsequent interventions, and comprehensively analyzes data before the marked cycles to propose intervention opportunities based on the lowest overall impact, thereby optimizing maintenance strategies.

[0093] By regularly checking various production parameters and determining whether the data exceeds the limit according to the preset cycle, we can implement phased analysis and dynamic adjustments to ensure the continuity and stability of production.

[0094] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A production risk identification system for a battery automatic assembly production line, characterized in that: include: Management module (1), used to connect to the bus system of the assembly production line and obtain data and instruction permissions; An assembly collection module (2) is used to collect the operating time, number of failures, downtime and operating efficiency of each assembly line equipment to form an operating data set; A quality inspection module (3) for acquiring and analyzing assembly product quality data during the production cycle in real time and generating a quality score; Supply access module (4), used to obtain and update the completion progress of production orders in real time; The assembly simulation module (5) is used to construct an assembly model that simulates the production impact coefficient of the current product in the assembly link by using the historical operation data, assembly progress, assembly product quality score and production order completion status of a certain equipment as samples; The impact prediction module (6) is used to obtain the trained assembly model through the assembly simulation module (5), input the current operation data of a certain equipment, assembly progress, assembly product quality score and production order completion status, and predict and output the production impact coefficient of the equipment on the production order completion status and assembly product quality score in the next several cycles; A determination module (7) is used to preset a standard production impact coefficient for the current cycle, and based on the standard production impact coefficient, to judge the production impact coefficients of several cycles output by the impact prediction module (6), and to mark the cycles to which the production impact coefficients exceed the preset standard production impact coefficients; A joint analysis unit (8) is used to comprehensively analyze several cycles before the cycle marked by the determination module (7), combine the current order status coefficient and the equipment life impact coefficient, and output a recommended cycle for starting intervention; The phase identification module (9) is used to perform real-time detection within the intervention period recommended by the joint analysis unit (8), obtain the actual expression of the production impact coefficient in stages according to the preset period, judge one by one whether the production impact coefficient of the corresponding period exceeds the preset limit, and create a smart contract on the blockchain to automatically trigger the contract terms adjustment intervention measures.

2. A production risk identification system for a battery automatic assembly line according to claim 1, characterized in that: The management module (1) is interactively connected to a case storage module (12) via a wireless network. The case storage module (12) is used to store historical cases of operating data of production-affecting parameters, assembly progress, assembly product quality scores, and production order completion status parameters through customized verification, and to mark and classify the stored historical cases as a sample library for the assembly simulation module (5).

3. The production risk identification system for a battery automatic assembly line according to claim 1 is characterized in that: The calculation formula of the production impact coefficient output by the impact prediction module (6) is: In the formula, P(t+i) represents the production impact coefficient in the time period t+i, T0 represents the intercept term of the model, and a j Represents the weight of the input variable, X j (t) represents the jth device operation data in time period t, n represents the number of input variables, b m Represents the weight of the assembly progress variable, U m (t) represents the progress coefficient of the current assembly work, P represents the number of assembly progress variables, c l represents the weight of the product quality score, d represents the weight of the production order status, Q l (t) represents the current production order completion status, A(t) represents the impact of equipment aging or wear on production capacity, C(t,i) represents the impact of equipment at time t on the future i cycles, V i represents the random influence coefficient related to the time period, e represents the weight of the equipment life influence coefficient, and f represents the weight of the interaction influence.

4. A production risk identification system for a battery automatic assembly line according to claim 3, characterized in that: The working logic of the calculation formula of the production impact coefficient is: Collect all input parameters related to the prediction of production impact coefficients for several future periods; Use regression analysis to calculate model parameters, and use the least squares method to minimize the error between the predicted value and the actual value to obtain the optimal parameters; The model parameters are determined, and the required input data are collected in each cycle to calculate the production impact coefficients for several future cycles.

5. The production risk identification system for a battery automatic assembly line according to claim 1, characterized in that: The joint analysis unit (8) is provided with submodules at a lower level, and the submodules include: an order extraction module (81), a machine analysis module (82), a cross-identification module (83), and an opinion output module (84). The order extraction module (81), the machine analysis module (82), and the cross-identification module (83) are interactively connected via a wireless network, and the cross-identification module (83) and the opinion output module (84) are interactively connected via a wireless network, wherein: An order extraction module (81) is used to extract order information of several periods before the period marked by the determination module (7) through the impact prediction module (6), including order quantity, delivery date and current completion progress, and output an order status coefficient; The machine analysis module (82) is used to obtain real-time equipment operation data and fault data through the assembly acquisition module (2), and to evaluate the equipment operation status for several cycles using pre-set performance indicators, and output the equipment life impact coefficient for several cycles before the cycle marked by the judgment module (7); A cross-identification module (83) is used to integrate the order status coefficients of several cycles provided by the order extraction module (81) and the equipment life impact coefficients of several cycles provided by the machine analysis module (82), compare the production impact coefficient of each cycle with the preset standard production impact coefficient, and identify the fluctuation state of the two types of coefficients; An opinion output module (84) is used to summarize the recognition results of the cross recognition module (83) in the fluctuation state of several cycles, sort them from high to low priority, and output a cycle with the highest priority as the cycle for recommended intervention; The cross-identification module (83) synchronously receives feedback data from the order extraction module (81) and the machine analysis module (82), and performs attribute matching of the two types of feedback data within the current cycle to form several data structures with the same attributes as initial integration references.

6. The production risk identification system for a battery automatic assembly line according to claim 1, characterized in that: The calculation formula for the cross-recognition module (83) to identify two types of coefficient fluctuation states is: Where R i Represents the order status coefficient OS in the i-th period i and equipment life impact coefficient EL i Relative to their respective target values OS target and EL target The relative difference in OS i Represents the order status coefficient of the i-th period, OS target Represents the target standard value of the order status coefficient, EL i Represents the equipment life impact coefficient of the i-th cycle, EL target Represents the target standard value of the equipment life impact coefficient; The calculation formula of the two types of coefficient fluctuation states is obtained by R i , perform statistical analysis and calculate the average value of the overall fluctuation state. If the average value exceeds a certain preset threshold, it is considered that there is an obvious fluctuation state.

7. A production risk identification system for a battery automatic assembly line according to claim 6, characterized in that: The working logic of creating a smart contract on the blockchain to intervene and automatically trigger the contract terms adjustment intervention measures in the stage identification module (9) is as follows: The contract terms set the monitoring period, production conditions and intervention conditions. During the phase change, when an actual production parameter exceeds the set intervention threshold, the smart contract will automatically trigger one or more intervention measures, including: Each stage change will generate an unalterable hash value through real-time recording and storage in the blockchain. Based on the actual expression data of the collected production impact coefficient, contextual data including the equipment's operating status, environmental parameters and quality inspection results will be generated; The smart contract deployed on the blockchain will be adjusted accordingly based on real-time contextual data, triggering different contract terms according to contextual changes. When a preset specific context is identified, the execution parameters of the smart contract will be dynamically changed based on custom rules and serve as an adjustment reference for the joint analysis unit (8).

8. The production risk identification system for a battery automatic assembly line according to claim 1, characterized in that: The determination module (7) and the stage identification module (9) are connected to an alarm module (10) via an electrical medium. The alarm module (10) is used to receive the determination results of the determination module (7) and the stage identification module (9) in real time. When any of the determination results is negative, an alarm behavior feedback is immediately provided to the management module (1).

9. The production risk identification system for a battery automatic assembly line according to claim 1, characterized in that: The stage identification module (9) is interactively connected to the patrol module (11) via a wireless network. The patrol module (11) is used to monitor in real time the change in the production impact coefficient after the intervention of the joint analysis unit (8), evaluate the identification results of the stage identification module (9), and synchronously feed back to the management module (1).

10. The production risk identification system for a battery automatic assembly line according to claim 1, characterized in that: The management module (1) is interactively connected to the assembly acquisition module (2), the quality inspection module (3) and the supply access module (4) via a wireless network; the assembly simulation module (5) is interactively connected to the assembly acquisition module (2), the quality inspection module (3), the supply access module (4) and the impact prediction module (6) via a wireless network; the impact prediction module (6) is interactively connected to the determination module (7) via a wireless network; the determination module (7) is interactively connected to the joint analysis unit (8) via a wireless network; and the joint analysis unit (8) is interactively connected to the stage identification module (9) via a wireless network.

Images